Format
Individual Presentation
First Presenter's Institution
Houston County Schools
Second Presenter's Institution
N/A
Third Presenter's Institution
N/A
Fourth Presenter's Institution
N/A
Fifth Presenter's Institution
N/A
Location
Scarbrough 1
Strand #1
Academic Achievement & School Leadership
Strand #2
Safety & Violence Prevention
Relevance
This presentation relates to the “head” strand as it describes the continued development of a risk prevention model, Houston At-Risk Profiles (HARP). HARP seeks to identify those students most at risk of failure and provides support in connecting those indicators with research-based interventions to improve academic achievement and reduce overall risk of failure. HARP is a result of a longitudinal analysis of over ten years of student performance data and represents a concerted effort to identify risk indicators across six distinct domains: reading, writing, English language, mathematics, behavior, and attendance.
Brief Program Description
This presentation chronicles the work of Houston County Schools in developing a risk prevention model built on more than ten years of longitudinal student data. In its second year of implementation, Houston At-Risk Profiles (HARP), has proven effective in identifying those students most in need of support and linking them to interventions and supports that lead to improved outcomes and significantly reduces the risk of failure.
Summary
Building a risk prevention model can seem like daunting task for any educator. In a market flooded with products built largely on data from other systems and providing support only from a sample model, Houston County sought to build a risk prevention model suited to the individual needs of students attending Houston County Schools. Using over ten years of longitudinal data warehoused in our student information system, machine learning algorithms written as a result of analyzing this data, we developed a six-domain risk prevention model. Our model, Houston At-Risk Profiles (HARP), helps teachers throughout our district identify students experiencing multiple risks, provides guidance on the specific risks students are experiencing, and then recommends interventions based on historical outcomes of students who experienced similar risk profiles. We learned a great deal during our first year of implementation and have re-tooled our model to include additional risk indicators and have subsequently seen a significant increase in the effectiveness of the model. Schools who analyzed and disseminated HARP data to teachers and other stakeholders saw an increase in overall achievement and saw a significant reduction in the number of and severity of risk indicators for students at the greatest risk of failure. We would like to share our journey, as well as practical advice for systems considering developing their own models, to show that risk prevention is possible and can be effective in improving achievement outcomes for all students at risk of failure.
Evidence
Of the 38 schools in our district, 11 used the HARP model with fidelity during the first full year of HARP implementation. At each of the 11 schools, close to 80% of the students experiencing the highest calculated risk of failure demonstrated improved proficiency on state standardized assessments. One elementary school saw a 40% increase in the percentage of students reading on grade level, and a middle school saw risk indicators fall for an overwhelming majority of students served under the model. By building our model on historical data, we were able to more effectively identify risk factors and intervene on behalf of students across the district. We are excited to see the results of our continued use of the model and look forward to its continued improvement as a result of these efforts. Our risk model can now more accurately predict which students will require support and provide timely intervention, targeted at specific deficits.
Biographical Sketch
Steven Hornyak serves as the System Intervention Specialist for Houston County Schools in Perry, Georgia. Steven began his career in education ten years ago, after a brief stint in the corporate world. Building on a background in Risk Management, Steven has worked to develop models to reduce the risk of failure for all students in Houston County, but has a special passion for helping to identify promising practices and interventions for students experiencing multiple risks.
Keyword Descriptors
Risk Prevention, Statistical Modeling, Risk Profiles, Analytics
Presentation Year
2018
Start Date
3-6-2018 1:15 PM
End Date
3-6-2018 2:00 PM
Recommended Citation
Hornyak, Steven, "Building a Better Risk Prevention Model" (2018). National Youth Advocacy and Resilience Conference. 102.
https://digitalcommons.georgiasouthern.edu/nyar_savannah/2018/2018/102
Included in
Analysis Commons, Applied Statistics Commons, Categorical Data Analysis Commons, Curriculum and Instruction Commons, Disability and Equity in Education Commons, Educational Assessment, Evaluation, and Research Commons, Educational Leadership Commons, Educational Methods Commons, Elementary and Middle and Secondary Education Administration Commons, Elementary Education Commons, Elementary Education and Teaching Commons, Junior High, Intermediate, Middle School Education and Teaching Commons, Longitudinal Data Analysis and Time Series Commons, Multivariate Analysis Commons, Other Educational Administration and Supervision Commons, Other Mathematics Commons, Other Statistics and Probability Commons, Other Teacher Education and Professional Development Commons, Secondary Education and Teaching Commons, Special Education Administration Commons, Special Education and Teaching Commons, Statistical Methodology Commons, Statistical Models Commons
Building a Better Risk Prevention Model
Scarbrough 1
This presentation chronicles the work of Houston County Schools in developing a risk prevention model built on more than ten years of longitudinal student data. In its second year of implementation, Houston At-Risk Profiles (HARP), has proven effective in identifying those students most in need of support and linking them to interventions and supports that lead to improved outcomes and significantly reduces the risk of failure.
